414 Liijirovtd Sltam-Engine w'l/lxul a Pi/loa. 



never requires fleam (Ironger than the atmofphere. From the efTeft of tliis engine under 

 circumftanccs of fuch advantage, it may fairly be concluded that tlic a£lion of ftcam againft 

 water in forcing can never be beneficial except at a place where fuel could be had ex- 

 tremely clicap. Thirdly, it was found at the firll conftruction of this engine, that the con- 

 fumption of fleam by contact with the water was fo great that it could not be worked with 

 the fmalleft advantage. This defeft was remedied in the prcfent engine, as well as in an- 

 other at Norwich, by fixing a fmall air-valve in the ileam-box, which was flruck for an in- 

 ftant immediately before the admiffion of the fleam. It may be prefurocd that the air oc- 

 cupied a fpace above the water, r.nd prevented their coming together. Mr. Kier, however, 

 is difpofcd to think that the efl^eft docs not take place in this manner, but by fomc mixture 

 and fuddcn dilatation of the two fluids ; becaufe he imagines the mifchicf from the wet 

 cylinder would be the fame upon defccnding fleam. Much, however, may be faid in defence 

 of the oppofite opinion. The air- valve is not at prefent ufed, becaufe the engine docs very 

 well without it •, which is fuppofed to arife from its being lefs air-tight than it was at firft. 

 Fourthly, the motion of the overfliot wheel is regulated by an apparatus called a governor, 

 invented, as I think, by Mr. Watt, and reprcfented in figure 4. The bar HI revolves by 

 communication with the engine, and carries round the balls A and B, which move on a 

 fixed ionifC. When the rotation is very quick, the balls fly out, and draw down the 

 points DE, and confequently F, which is fo conftrui5le<l as to flide upon the bar or axis HI. 

 A lever, FG, conncifled with the fluice of the upper ciftern GG, fig. 2. is therefore made 

 to fall or rife accordingly as the velocity is greater or lefs. By this difpofilion, when the 

 ■wheel moves very fpeedily, from lightncfs of work or any oth;r caufc, the quantity of 

 water thrown down from the upper ciftern is immediately diminilhed.4 and, contr.irywife, 

 the quantity of water is rendered greater when the flownefs of the movement fliews that it 

 is wanted. When I faw the engine at work, it had but one cleat and block upon its wheel, 

 and two men were at work at flrong lathes, roughing out certain pieces of iron about two 

 inches in diameter. Whether it were the unfteadincfs of this work, or the want of fnf- 

 ficient celerity of communication, I know not: hut the variations in this governor were 

 upon the whole more confidcraUe than I have remarked in a lever-engine with a fly and 

 this apparatus alfo. Fifthly, as the inje£lion-water makes a conflant addition to the fum 

 of the water contained in the cifterns NO and GG, the furplus is allowed to overflow. If 

 it were to overflow the lower ciftern, it would warm the water in the inje£lion well ; for 

 which reafon a contrivance becomes neceflary to prevent that cflefl. Mr. Kier's contrivance 

 is this : — The water of the upper ciftern, when at a certain elevation, overflows into a com- 

 munication which conveys the water to the lower ; for which reafon its contents can never 

 rife above a certain level. There is alfo a pipe from the bottom of the upper ciftern, 

 which is recurved upwards to communicate with a gutter a little lower than the place of 

 overflowing juft mentioned. It is clear, therefore, that the upper cillern would always over- 

 flow at this pipe fooner than at the place of communication with the lower ciftern, if there 

 were not a valve within the upper ciftern, over the mouth of the recurved pipe. Now it 

 is fo managed by means of a float in the lower ciftern, that this valve in the upper ftiall be 

 opened whenever the water below rifes to a certain level. And confequently, as foon as ever 

 the lower ciftern has received this portion, it can receive no more, becaufe all the fubfe- 

 quent addition of water will pafs away by the recurved pipe. 



This 



